Solar Panel Shadow Simulation — 3D Web Application

🔍 The Challenge

Simulating solar shadows and irradiance across large photovoltaic fields is computationally demanding. Each simulation depends on numerous parameters — geographic position, date, solar trajectory, panel orientation, tilt, and spacing — requiring real-time 3D feedback to guide configuration choices. The goal was to deliver a responsive and accurate web app capable of handling heavy physics calculations while remaining fluid and accessible to non-technical users.

• ⚙️ High-precision irradiance and shadow modeling
• 🌍 Parameter-rich simulations (geometry, tilt, azimuth, date…)
• 🧮 Heavy computational load across large panel arrays
• 💻 Need for smooth visualization and background computation

💡 The Solution

The Solar Panel Shadow Simulation App provides a complete, parameter-driven environment for modeling and visualizing solar installations. Users can define every aspect of a setup — from panel dimensions and pitch to site latitude and time resolution — then launch simulations that compute irradiance, shadow overlap, and light diffusion using the open-source scientific library PASE.

The results are displayed in an interactive 3D visualization window, enabling users to view how shadows evolve through the day and how total irradiance changes with geometry. Computations are handled in the background by a dedicated backend service, allowing the user to close the page or run multiple simulations in parallel.

⚙️ Technical Architecture

• Frontend: Streamlit web interface with 3D scene rendering and parameter control
• Backend: Python API handling simulation requests asynchronously
• Computation: Multi-process system using multiprocessing to run concurrent simulations
• Scientific engine: PASE library for solar geometry and irradiance computation
• Containerization: Full Docker deployment with persistent volume management
• Task management: Background queue system allowing simulations to continue when UI is closed

This architecture ensures efficiency, scalability, and robustness, making it suitable for large solar projects or educational demos requiring multiple concurrent simulations.

📊 Results

The application enables engineers and clients to visualize and quantify the impact of configuration changes instantly. The backend’s multiprocessing system reduces total simulation time by a lot (the more thread the faster) compared to sequential execution.

• ☀️ Realistic 3D irradiance and shadow modeling
• 🚀 Parallel execution of multiple simulations
• 🧩 Modular architecture — parameters easily extendable
• 💾 Results saved for offline analysis
• 📈 Improved DevOps workflow and deployment automation

🧠 My Role

I was responsible for developing the backend architecture and implementing the multi-process computation system. My work focused on performance, scalability, and seamless integration with the Streamlit interface.

• ⚙️ Designed and implemented the multi-process simulation backend
• 🐳 Set up full Dockerized environment and CI/CD deployment
• 🧮 Integrated PASE scientific library for shadow and irradiance modeling
• 🔗 Developed communication API between backend and Streamlit frontend
• 🧠 Optimized resource allocation and background task scheduling

This project reinforced my expertise in Python system design, DevOps, multiprocessing, and scientific software integration — bridging simulation performance with web-based usability.